User Portlet
Featured Contributor
| Discussions |
|---|
| Isn't Correlation just reversing f[t] aka f[-t] and performing Convolution. Convolve[f[-x],g[x],x,y] |
| If we consider readability I would prefer: ({x,y} |-> x^2 + y^2)[3, 2] versus: (#1^2 + #2^2)&[3, 2] |
| Total does more than simple summation. With Total you can control at which nesting level the summation happens. |
| Unhappy with existing ellipse fitting I present a function based on Linear Least Squares and SVD, the latter symbolically resolved and not directly used/performed for the actual fit calculation. &[Wolfram Notebook][1] [1]:... |
| try both and report your experience ;) |
| You are not talking about the discrete convolution. In Mathematica the discrete convolution is performed by ListConvolve. You might try to tabulate your functions over some discrete x / y region an then in deed perform a discrete convolution via... |
| Sqrt[(a b c + d e g)^2] // PowerExpand |
| A little bit late but nevertheless: FourierShift[a_] := RotateRight[a, Quotient[Dimensions@a - 1, 2]] |
| Nobody questioned that this works. |
| Hi Ehud What you are trying is called side effect programming which is considered as bad style and should be generally avoided in any programming. In partikular you are trying to pass a parameter by reference which is not the "Mathematica Way" . ... |